no code implementations • 24 Sep 2023 • Can Peng, Piotr Koniusz, Kaiyu Guo, Brian C. Lovell, Peyman Moghadam
Deep learning models suffer from catastrophic forgetting when being fine-tuned with samples of new classes.
no code implementations • 20 Dec 2022 • Meng Li, Chaoyi Li, Can Peng, Brian Lovell
Extensive experiments on the histopathology datasets show that leveraging our synthetic augmentation framework results in significant and consistent improvements in classification performance.
1 code implementation • 30 Jul 2022 • Can Peng, Kun Zhao, Tianren Wang, Meng Li, Brian C. Lovell
The continual appearance of new objects in the visual world poses considerable challenges for current deep learning methods in real-world deployments.
no code implementations • 8 Sep 2021 • Tianren Wang, Can Peng, Teng Zhang, Brian Lovell
With the excellent disentanglement properties of state-of-the-art generative models, image editing has been the dominant approach to control the attributes of synthesised face images.
no code implementations • 12 Aug 2021 • Can Peng, Kun Zhao, Sam Maksoud, Tianren Wang, Brian C. Lovell
In this paper, we aim to alleviate this performance decay on multi-step incremental detection tasks by proposing a dilatable incremental object detector (DIODE).
no code implementations • 19 Apr 2021 • Sam Maksoud, Kun Zhao, Can Peng, Brian C. Lovell
To address this problem we present a method for performing BDL, namely Kernel Seed Networks (KSN), which does not require a 2-fold increase in the number of parameters.
no code implementations • 31 Dec 2020 • Can Peng, Kun Zhao, Sam Maksoud, Meng Li, Brian C. Lovell
Incremental learning requires a model to continually learn new tasks from streaming data.
1 code implementation • 9 Mar 2020 • Can Peng, Kun Zhao, Brian C. Lovell
To address this problem, incremental learning methods have been explored which preserve the old knowledge of deep learning models.
no code implementations • 22 Sep 2019 • Can Peng, Kun Zhao, Arnold Wiliem, Teng Zhang, Peter Hobson, Anthony Jennings, Brian C. Lovell
Critical findings are observed: (1) The best balance between detection accuracy, detection speed and file size is achieved at 8 times downsampling captured with a $40\times$ objective; (2) compression which reduces the file size dramatically, does not necessarily have an adverse effect on overall accuracy; (3) reducing the amount of training data to some extents causes a drop in precision but has a negligible impact on the recall; (4) in most cases, Faster R-CNN achieves the best accuracy in the glomerulus detection task.